Vmware, Inc.

Hewlett-packard Development Company, L.p.

Ritesh Ranjan Jha

Nikhil Jaiswal

Chandrashekhar Jha

Jobin Raju George

Pushkar Patil

Yash Bhatnagar

Vaidic Joshi

Kumaravel Ganesan

Mageshwaran Rajendran

Shivam Satija

Hemani Katyal

Soumya Panigrahi

Rajesh Bhatia

Suchit Dhakate

Puneet Kumar Arora

James Allely Rowson

Shubham Verma

Rachil Chandran

Venugopal Kumarahalli Srinivasmurthy

Dattathreya Sathyamurthy

Karthik Seshadri

Akshay Bhandari

Rajat Garg

Siddartha Laxman Karibhimanvar

Aayush Asawa

Swarnalatha Pasupuleti

Lenin Fernandes

Vishwanath Ramaiah Nanjundaiah

Radhakrishnan Devarajan

Narayanasamy Ramesh

Priyank Agarwal

Vishwanath Rn

Chaitanya Krishna Mullangi

Navya Sree Tirunagari

Jasmine Ejner

Susobhit Panigrahi

Sundharam Balathirupurasundari

Shivanju Awasthi

Dhanendra Jain

Gurpreet Singh Bhatia

Balathirupurasundari Sundharam

The current document is directed to methods and systems that automatically identify log/event-message traces and computational-entity transactions within collections and/or streams of log/event messages. Automated identification of log/event-message traces provides the basis for automated interpretation, by automated computer-system administration-and-the management subsystems, of the information represented by collections and/or streams of log/event messages. Disclosed approaches to automatically identifying log/event-message traces and computational-entity involve identifying log/event-message types, generating time-series-like log/event-message-type occurrence signals from log/event-message collections and/or streams, and computing cross correlations between pairs of log/event-message-type occurrence signals. In one implementation, a strongly-correlated-type graph is generated from the computed cross correlations, from which connected-components subgraphs, corresponding to computational-entity transactions, are extracted. Log/event-message traces are then extracted from acyclic graphs generated from the connected-component subgraphs.

Methods and systems that identify computational-entity transactions and corresponding log/event-message traces from streams and/or collections of log/event messages

An example system includes memory, programmable circuitry, and machine readable instructions to program the programmable circuitry to: obtain utilization metric information corresponding to utilization metrics collected over a time interval, the utilization metrics corresponding to allocated resources utilized by containers, the containers associated with a cluster, obtain a request to generate priority classes for the containers in the cluster, the priority classes indicative of which containers have a greater priority in the cluster, and generate the priority classes for the containers based on the utilization metric information and a count of network interactions corresponding to the containers for the time interval.

Methods and apparatus to determine container priorities in virtualized computing environments

Methods and systems described herein are directed to aggregating and querying log messages. Methods and systems determine event types of log message generated by event sources of the distributed computing system. The event types are aggregated into aggregated records for a shortest time unit and event types are aggregated into aggregated records for longer time units based on the aggregated records associated with the shortest time unit. In response to a query regarding occurrences of an event type in a query time interval, the query time interval is split into subintervals with time lengths that range from the shortest time unit to a longest time unit that lie within the query time interval. The method determines a total event count of occurrences of the event type in the query time interval based on the aggregated records with time stamps in the subintervals. The event count in the query time interval may be used to detect abnormal behavior of the event sources.

Methods and systems for aggregating and querying log messages

The current document is directed to methods and systems that efficiently process and store log/event messages generated within distributed computer facilities. Various different types of initial processing steps may be applied to a stream of log/event messages received by a message-collector system or a message-ingestion-and-processing subsystem. The currently disclosed methods and systems employ additional pre-processing steps to identify the types of received log/event messages, monitor event-type-associated log/event-message-usage-delay histories, and employ time-series-analysis-based and/or neural-network-based estimation of event-type-associated log/event-message usage to efficiently store log/event-messages in low-cost and low-latency storage facilities.

Method and subsystem within a distributed log-analytics system that automatically determines and enforces log-retention periods for received log-event messages

This disclosure describes how data supporting real-time reporting services can be cached during a log intake process. In particular, instead of caching all the log data being generated by an operational system, only the log data relevant to existing queries associated with the real-time reporting services are cached. In some embodiments, only particular metrics contained within the log data are stored for rapid access by the real-time reporting services.

Real-time dashboards, alerts and analytics for a log intelligence system

The current document is directed to methods and systems that efficiently process log/event messages within and among distributed computer facilities. Various different types of initial processing steps may be applied to a stream of log/event messages received by a message-collector system or a message-ingestion-and-processing system. By including a pre-processing step two identify the type of a received log/event message, and by specifying initial-processing-step criteria with respect to log/event-message types, significant increases in the efficiency of log/event-message preprocessing by message-collector systems and message-ingestion-and-processing systems is achieved.

Efficient event-type-based log/event-message processing in a distributed log-analytics system

The current document is directed to methods and systems that sample log/event messages for downstream processing by log/event-message systems incorporated within distributed computer facilities. The data-collection, data-storage, and data-querying functionalities of log/event-message systems provide a basis for distributed log-analytics systems which, in turn, provide a basis for automated and semi-automated system-administration-and-management systems. By sampling log/event-messages, rather than processing and storing every log/event-message generated within a distributed computer system, a log/event-message system significantly decreases data-storage-capacity, computational-bandwidth, and networking-bandwidth overheads involved in processing and retaining large numbers of log/event messages that do not provide sufficient useful information to justify these costs. Increase in efficiencies of log/event-message systems obtained by sampling translate directly into increases in bandwidths of distributed computer systems, in general, and to increases in time periods during which useful log/event messages can be stored.

Methods and systems that sample log/event messages in a distributed log-analytics system

The current document is directed to methods and systems that efficiently cache log/event messages in log/event-message systems incorporated within distributed computer facilities. The log/event-message systems provide the data-collection, data-storage, and data-querying functionalities that provide a basis for distributed log-analytics systems which, in turn, provide a basis for automated and semi-automated system-administration-and-management systems. Efficient caching significantly extends the period of time that networking failures and log/event-message processing delays can be tolerated by log/event-message systems without significant loss of information. Rather than caching individual log/event messages, message collectors that employ the currently disclosed technologies cache information regarding log/event-message types, leading to efficient compression of the cached information. As a result, message collectors can cache log/event messages without significant loss of information for time periods of several orders of magnitude greater than those achieved by caching individual log/event messages.

Methods and systems that efficiently cache log/event messages in a distributed log-analytics system

An example apparatus includes memory, and at least one processor to execute instructions to assign first containers to a first cluster and second containers to a second cluster based on the first containers including first allocated resources that satisfy a first threshold number of allocated resources and the second containers including second allocated resources that satisfy a second threshold number of allocated resources, determine a representative interaction count value for a first one of the first containers, the representative interaction count value based on a first network interaction metric corresponding to an interaction between the first one of the first containers and a combination of at least one of the first containers and at least one of the second containers, and generate a priority class for the first one of the first containers based on the representative interaction count value.

Disclosed are various embodiments for automating the prediction of workload tenures in datacenter environments. In some embodiments, parameters are identified for a plurality of workloads of a software defined data center. A machine learning model is trained to determine a predicted tenure based on parameters of the workloads. A workload for the software defined data center is configured to include at least one workload parameter. The workload is processed using the trained machine learning model to determine the predicted tenure. An input to the machine learning model includes the at least one workload parameter.

Workload tenure prediction for capacity planning

Methods and systems that automatically rank log/event messages and log/event-message transactions to facilitate analysis of log/event-messages generated within distributed-computer systems are disclosed. A base-window dataset and current-window dataset are selected for diagnosis of a particular error or failure and processed to generate a transaction sequence for each dataset corresponding to log/event-message traces identified in the datasets. Then, frequencies of occurrence of log/event-message types relative to transaction types are generated for each dataset. From these two sets of relative frequencies of occurrence, changes in the relative frequency of occurrence for each log/event-message-type/transaction-type pair are generated. Normalized scores for log/event-message-type/transaction-type pairs and scores for transaction types are then generated from the changes in the relative frequency of occurrence. The generated scores reflect the relevance of log/event-messages in traces corresponding to particular transaction as well as the relevance of transaction types to the error or failure.

Methods and systems that rank and display log/event messages and transactions

A system and method for sampling application programming interface (API) execution traces in a computer system uses feature vectors of the API execution traces that are generated using trace-context information. The feature vectors are then used to group the API execution traces into clusters. For the cluster, sampling rates are generated so that a sampling rate is assigned to each of the clusters. The sampling rates are then applied to the API execution traces to adaptively sample the API execution traces based on the clusters to which the API execution traces belong.

System and method for adaptively sampling application programming interface execution traces based on clustering

The present disclosure provides an approach for training a machine learning model by first training the model on a generic dataset and then iteratively training the model on 'easy' domain specific training data before moving on to 'difficult' domain specific training data. Inputs of a domain-specific dataset are run on the generically-trained model to determine which inputs generate an accuracy score above a threshold. The inputs with an accuracy score above a threshold are used to retrain the model, along with the corresponding outputs. The retraining continues until all domain specific dataset has been used to train the model, or until no remaining inputs of the domain specific dataset generate an accuracy score, when run on the model, that is above a threshold.

Enhanced learning with feedback loop for machine reading comprehension models

System and computer-implemented method for constructing a topology for an application distributed over multiple virtual computing instances in which a set of resource utilization metrics are collected for multiple computing instances. A propagation sequence caused by an execution of the application is determined based on the set of resource utilization metrics. A graph may be created with the propagation sequence, wherein each virtual computing instance in the multiple computing instances is a possible node in the graph and the graph includes all nodes associated with the application.

Method and apparatus for non-intrusive agentless platform-agnostic application topology discovery

System and computer-implemented method for managing workloads in a computing environment use collected resource utilization metrics for a workload of a compute instance running in the computing environment that are superimposed to produce a unified time series of the resource utilization metrics for the workload. Undesired periodic patterns are then removed from the unified time series of the resource utilization metrics, which is used to identify the compute instance as a particular type of compute instances. Based on the identification, an action is recommended and implemented on the compute instance.

Growing community of inventors

Karnataka, India

Ritesh Ranjan Jha